JetsonPDF.Reader 1.0.0

JetsonPDF.Reader

Parses PDF files into a navigable, in-memory object model — ReadDocument — following ISO 32000-2 (PDF 2.0). Headless, no native deps, pure net8.0.

using JetsonPDF.Reading;

ReadDocument doc = Reader.Load("input.pdf");

Console.WriteLine($"{doc.Pages.Count} pages, version {doc.Version}");
foreach (var page in doc.Pages)
{
    Console.WriteLine($"  page {page.Index + 1}: {page.MediaBox.Width}×{page.MediaBox.Height}");
    foreach (var item in page.Items)
        // text runs, vector paths, images, inline images, shadings, …
        Console.WriteLine($"    {item.GetType().Name}");
}

Reader.Load accepts a path, Stream, or byte[], with an optional password for encrypted documents.

What's covered

• File structure: classic xref + xref streams, object streams, encrypted files (RC4-40/128, AES-128, AES-256), incremental updates, linearized ("Fast Web View") layouts, hint tables.
• Content streams: text, vector paths, raster images, inline images (BI / ID / EI), Form XObjects, clipping, marked content, optional content.
• Color: Device + CIE-based (CalGray, CalRGB, Lab), ICC, Indexed, Separation, DeviceN/NChannel, Pattern; full PdfFunction evaluator (Types 0, 2, 3, 4).
• Shadings: axial, radial, function-based, mesh (Types 4–7 decoded to triangle / patch lists).
• Fonts: Type 1 with /Differences via Adobe Glyph List, TrueType / OpenType, CID composite (Identity-H), Type 3 with /ToUnicode preference.
• Filters: Flate (with PNG predictors), LZW, ASCII85, ASCIIHex, RunLength, CCITTFax (T.4 1D + T.4 2D / K>0 + T.6), JBIG2 (MMR + arithmetic + symbol dicts + text regions + refinement + halftones + pattern dicts + Huffman tables), JPX (JP2 metadata extraction; image data passthrough).
• Annotations: Link, Highlight / Underline / StrikeOut / Squiggly, FreeText, Stamp, Caret, FileAttachment, Popup, Sound, Redact, Line / Polygon / PolyLine, Ink, plus markup target binding.
• Forms: AcroForm widgets (Tx / Btn / Ch / Sig / barcode), /AA triggers, /CO calculation order, signature /SV seed values, /Lock, DocMDP / FieldMDP, DocTimeStamp, DSS / VRI for LTV.
• Document metadata: Outlines, page labels, named destinations, viewer preferences, page layout / mode, OpenAction, XMP packet, OCG / layers, associated files, structure tree (real nested elements + role/class maps), PDF/A-1b/2/3 + PDF/UA-1/2 conformance flags, output intents, /Measure + /VP viewports + coordinate systems.

Optional validation

ReadDocument doc = Reader.Load("input.pdf");
var issues = doc.Validate(); // PDF/A-1b, PDF/A-2/3 a/u/b, PDF/UA-1/2 rules
foreach (var i in issues) Console.WriteLine($"{i.Severity}: {i.Code} {i.Message}");

Targets

• net8.0

License

MIT.